Skip to content

Low-rank matrix completion using Iterative Singular Value Thresholding and Alternating Minimization for predicting product ratings on Amazon datasets.

Notifications You must be signed in to change notification settings

amitkp57/personalized-product-recommendation

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

9 Commits
 
 
 
 
 
 
 
 

Repository files navigation

Venus

Low Rank Matrix Completion

Matrix Completion is the task of filling the missing elements in a matrix. The famous example is Netflix challenge: Given a rating matrix in which (i, j) element refers to rating for ith movie by jth user. There are a lot of missing entries in the matrix as a particular user is expected to rate only a few of the movies. So matrix completion is used in such cases to fill out the missing entries. Netflix used matrix completion to recommend new movies to the users. I used matrix completion on Amazon product ratings data to predict the missing ratings by the users. The below two algorithms are implemented for low-rank matrix completion:

  1. Iterative Singular Value Thresholding
  2. Alternative Minimization

About

Low-rank matrix completion using Iterative Singular Value Thresholding and Alternating Minimization for predicting product ratings on Amazon datasets.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages